It looks Romain was referring to failing early or late because of spotless (bind the check goal to validate or verify phase) Perso, I prefer to fail early (check at validate phase) because I find it frustrating to run some 10 minutes build (because -Prun-its) and then fail after so long because of a missing space :) And have to re-run everything again.... (more time wasted) But at the end of the day, it's very similar. It's just some personal preferences that I find wasting more time when minor/useless problems are detected later.
On Mon, 2 Dec 2024 at 08:35, Guillaume Nodet <gno...@apache.org> wrote: > > Currently, the build will fail both locally and on CI if your sources are > not formatted properly. > The proposal is to change local builds to auto-format and keep CI to a a > check. > > So it cannot slow your dev workflow, quite the opposite, as instead of > failing your build locally, the sources will be automatically formatted and > the build will succeed. Am I missing something ? > > Guillaume > > Le dim. 1 déc. 2024 à 16:37, Romain Manni-Bucau <rmannibu...@gmail.com> a > écrit : > > > Not sure it is that important since at the end it will fail but from my > > experience, bypassing it during the dev (ie being bound late) is nicer > > cause you iterate faster. > > That said we can PR to make it "failAtEnd", would be good anyway > > > > Romain Manni-Bucau > > @rmannibucau <https://x.com/rmannibucau> | Blog > > <https://rmannibucau.metawerx.net/> | Old Blog > > <http://rmannibucau.wordpress.com> | Github < > > https://github.com/rmannibucau> | > > LinkedIn <https://www.linkedin.com/in/rmannibucau> | Book > > < > > https://www.packtpub.com/en-us/product/java-ee-8-high-performance-9781788473064 > > > > > > > > > Le dim. 1 déc. 2024 à 13:00, Maarten Mulders <mthmuld...@apache.org> a > > écrit : > > > > > +1 to make it part of the Maven build, either in validate or in verify > > > phase. I usually do this in work projects, too. As Eliotte says, it > > > avoids breaking the build in CI. Preferably enabled by default, also on > > > local builds. > > > > > > > > > The only thing where I doubt is whether validate or verify would be a > > > good phase. > > > > > > Reasons for validate: early feedback on formatting, avoids compilation, > > > testing etc. if the code is not formatted correctly. > > > Reasons for verify: the important thing is that the code works, > > > formatting is the last concern. "First make it work, then make it > > pretty". > > > > > > > > > WDYT? > > > > > > > > > Maarten > > > > > > On 29/11/2024 13:27, Elliotte Rusty Harold wrote: > > > > While working on a third party plugin I noticed it was configured to > > > > autorun spotless as part of "mvn test" (and probably other goals): > > > > > > > > [INFO] --- spotless:2.43.0:apply (default) @ spotbugs-maven-plugin --- > > > > [INFO] Spotless.Format is keeping 10 files clean - 0 were changed to > > > > be clean, 0 were already clean, 10 were skipped because caching > > > > determined they were already clean > > > > [INFO] Spotless.Format is keeping 182 files clean - 0 were changed to > > > > be clean, 1 were already clean, 181 were skipped because caching > > > > determined they were already clean > > > > [INFO] > > > > > > > > We should do that too in our poms to avoid the recurring and annoying > > > > problems of forgetting to run spotless manually and consequently > > > > having a meaningless break in the CI that needs to be investigated. > > > > Integrated over all of us this should save a few days of developer > > > > time per year. > > > > > > > > I'd file a bug, but the way we've set up issue tracking this would be > > > > one per repo, and I'm disinclined to file the same bug 40 times. Let's > > > > just do it. > > > > > > > > > > > > > --------------------------------------------------------------------- > > > To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org > > > For additional commands, e-mail: dev-h...@maven.apache.org > > > > > > > > > > > -- > ------------------------ > Guillaume Nodet --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org For additional commands, e-mail: dev-h...@maven.apache.org
